1. Volume 10, Issue 8, Pages 695-704 (August 2003)
Irreversible Inhibition of CD13/Aminopeptidase N by the Antiangiogenic Agent Curcumin 
Joong Sup Shim, Jin Hee Kim, Hyun Young Cho, Young Na Yum, Seung Hee Kim, Hyun-Ju Park, Bum Sang Shim, Seung Hoon Choi, Ho Jeong Kwon 
Chemistry & Biology 
Volume 10, Issue 8, Pages (August 2003)
DOI: /S (03)

2. Figure 1 Structures of Compounds
(A) Curcumin and demethoxycurcumin were purified from commercial curcuminoids using thin layer chromatography.
(B) Hydrazinocurcumin is a new synthetic derivative of curcumin.
(C) Bestatin is a competitive inhibitor of APN.
Chemistry & Biology  , DOI: ( /S (03) )

3. Figure 2 The Effect of Curcumin Derivatives on APN Activity
(A) In vitro APN assay using purified aminopeptidase N.
(B) Kinetic analysis of enzyme inhibition by curcumin using Lineweaver-Burk plot of APN. Dixon plot was used to determine the Ki value (Ki = 11.2 μM).
(C and D) In vivo enzyme assay of APN in HUVECs (C) and in HT1080 (D) cells, respectively. Each value represents mean ± SE from three independent experiments. Cur, curcumin; D-Cur, demethoxycurcumin; H-Cur, hydrazinocurcumin; Best, bestatin.
Chemistry & Biology  , DOI: ( /S (03) )

4. Figure 3
Reversibility and Kinetic Analysis of APN Inhibition by Curcumin
(A) Reversibility of enzyme inhibition. Purified porcine APN was incubated with 0.5% methanol (Control), 10 μM bestatin (Best), and 10 μM curcumin (Cur) for 2 hr. The enzyme reaction solutions were filtered using Microcon-YM30, and the activity of the residual enzyme resuspended in PBS was determined. The white bars represent the enzymatic activity of APN before the filtration. The hatched and black bars represent the enzymatic activity of APN after the filtration one and three times, respectively. The APN activity of drug-untreated control was normalized to 100%, and drug-treated enzyme activity was expressed as a percentage of inhibition versus control. The data represent mean ± SE from three independent experiments. *, p < versus control APN activity (before filtration); #, p < versus control APN activity (after filtration three times).
(B) Kinetics analysis of enzyme inhibition. The enzymatic activity of APN was determined as described in Experimental Procedures. The kinetics of enzyme inhibition by bestatin (Best) or curcumin (Cur) is presented as an EV plot of initial velocity versus enzyme concentrations.
Chemistry & Biology  , DOI: ( /S (03) )

5. Figure 4
Analysis of the Interaction Between Curcumin and APN In Vitro and In Vivo
(A) The binding sensorgram of the curcumin and hydrazinocurcumin on APN from BIAcore analysis. Sensorgrams were obtained for 4 μM (bold dotted line), 16 μM (bold broken line), and 50 μM (bold line) of curcumin, 100 μM of hydrazinocurcumin (thin broken line), and 5% DMSO in flow buffer (thin line) against APN immobilized on the sensor chip.
(B) The plot of steady-state binding level (Req) against the concentration of curcumin. Each value was obtained from the equation as described in Experimental Procedures.
(C) FACS analysis of the competitive binding of curcumin with APN-specific antibody (WM15) to HUVECs. The curves indicated by white arrows represent HUVECs incubated with FITC-conjugated anti-mouse IgG only. Black arrows indicate the curves for HUVECs labeled with WM15 followed by FITC-conjugated anti-mouse IgG. Curves were obtained from antibody-labeled HUVECs incubated with 0 μM (a), 10 μM (b), and 50 μM (c) of curcumin and 50 μM of hydrazinocurcumin (d).
(D) The binding of WM15 to APN in either HT1080 (APN-positive) or MDA-MB-231 (APN-negative) cells. Note the increased fluorescence in HT1080 (a) but not in MDA-MB-231 cells (b).
Chemistry & Biology  , DOI: ( /S (03) )

6. Figure 5
Inhibition of Angiogenesis Activities by Curcumin and Bestatin
(A) Time and dose responses of curcumin on HUVECs. HUVECs were treated with curcumin, and cell growth was determined at various time points. The absorbance of MTT-formazan (540 nm, y axis) represents cell growth. Each value represents mean ± SE from three independent experiments.
(B) The effects of curcumin and bestatin on the bFGF-induced invasion of HUVECs. Serum-starved HUVECs in serum-free medium (SF) or treated with bFGF in the presence or absence of inhibitors were used for invasion assay. The invasiveness of cells in SF was used as a normalization control (100%). Each value represents the mean ± SE from three independent experiments. *, p < 0.03; **, p < 0.007; and #, p < 0.02 versus bFGF control.
(C) The effects of curcumin and bestatin on capillary tube formation of HUVECs. HUVECs with 30 ng/ml of bFGF (a), bFGF and 5 μM curcumin (b), bFGF and 10 μM curcumin (c), and bFGF and 100 μM bestatin (d). White arrows indicate the inhibition of tube networks by the agents.
(D) Inhibition of angiogenesis by curcumin in the CAM assay. Retinoic acid (RA, 1 μg/egg) and curcumin (Cur, 10 μg/egg) were applied to the CAM assay, and the inhibition ratio was calculated based on the percentage of angiogenic eggs to total numbers of eggs tested.
Chemistry & Biology  , DOI: ( /S (03) )

7. Figure 6 Analysis of APN Expression and Tumor Invasion
(A) RT-PCR analysis of APN expression in HUVECs. HUVECs were stimulated by bFGF (30 ng/ml), and RNA was isolated at the indicated time points. The isolated RNA was converted to cDNA in the presence or absence (No RT) of reverse transcriptase. Standard RT-PCR was performed using the primer pair specific for APN with or without (No Temp) cDNA template. The level of β-actin mRNA was used as an internal control. M denotes size marker.
(B) Quantitative data for RT-PCR results. The expression level of APN mRNA was determined by densitometry. Each value represents mean ± SE from three independent experiments. *, p < 0.04; **, p < versus bFGF (0 hr).
(C) APN expression in cultured tumor cell lines using RT-PCR with primers common to murine and human APN. The level of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA was used as an internal control.
(D) Relative invasiveness of various tumor cell lines. The invasiveness of each cell line was determined by the ratio of the number of invaded cells to total inoculated cells. Relative invasiveness of each cell line was calculated by the comparison with the invasiveness of HT1080 cells that were used as a normalization control (100%). Tumor invasion assay was performed as described in Experimental Procedures. Each value represents mean ± SE from three independent experiments. *, p < ; #, p < versus HT1080 control.
Chemistry & Biology  , DOI: ( /S (03) )

8. Figure 7
Curcumin Inhibits Tumor Invasion of APN-Positive Cells but Does Not Affect an APN-Negative Cell Line
(A) Effect of curcumin on C8161 and MDA-MB-231 cell invasion. The number of cells that invaded was determined using a microscope, and the data are presented as the average number of invaded cells. The invasiveness of each cell line without curcumin treatment was used as a normalization control (100%). *, p < versus C8161 control; #, p < versus MDA-MB-231 cells treated with 10 μM curcumin.
(B) Effects of curcumin and bestatin on bFGF-induced tumor invasion of three APN-positive cell lines. The invasiveness of each cell line in serum-free media was used as a normalization control (100%). Each value represents mean ± SE from three independent experiments. *, p < ; #, p < versus bFGF-stimulated HT1080 control.
Chemistry & Biology  , DOI: ( /S (03) )